windows workload: many ept_violation and mmio exits
I am running a windows workload which has 26 windows VMs running many instances of a J2EE workload. There are 13 pairs of an application server VM and database server VM. There seem to be quite a bit of vm_exits, and it looks over a third of them are mmio_exit:
Quoted 24 lines
efer_relo 0 exits 337139 fpu_reloa 247321 halt_exit 19092 halt_wake 18611 host_stat 247332 hypercall 0 insn_emul 184265 insn_emul 184265 invlpg 0 io_exits 69184 irq_exits 52953 irq_injec 48115 irq_windo 2411 largepage 19 mmio_exit 123554 mmu_cache 0 mmu_flood 0 mmu_pde_z 0 mmu_pte_u 0 mmu_pte_w 0 mmu_recyc 0 mmu_shado 0 mmu_unsyn 0
Quoted 8 lines
nmi_injec 0 nmi_windo 0 pf_fixed 19 pf_guest 0 remote_tl 0 request_i 0 signal_ex 0 tlb_flush 0
I collected a kvmtrace, and below is a very small portion of that. Is there a way I can figure out what device the mmio's are for? Also, is it normal to have lots of ept_violations? This is a 2 socket Nehalem system with SMT on.
Quoted 127 lines
qemu-system-x86-19673 [014] 213577.939614:
kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939624:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939624:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939627:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939629:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f214dqemu-system-x86-19673 [014] 213577.939631:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939633:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939634:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939636:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19332 [008] 213577.939637:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939638:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f24e2qemu-system-x86-19673 [014] 213577.939640:kvm_entry: vcpu 0
qemu-system-x86-19211 [010] 213577.939663:kvm_set_irq: gsi 11 level 1 source 0 qemu-system-x86-19211 [010] 213577.939664:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19211 [010] 213577.939665:kvm_apic_accept_irq: apicid 0 vec 130 (LowPrio|level) qemu-system-x86-19211 [010] 213577.939666:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19673 [014] 213577.939692:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939693:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939696:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19332 [008] 213577.939699:kvm_exit: reasonept_violationrip0xfffff80001b3af8eqemu-system-x86-19332 [008] 213577.939700:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939702:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f3da6qemu-system-x86-19563 [010] 213577.939702:kvm_set_irq: gsi 11 level 1 source 0 qemu-system-x86-19563 [010] 213577.939703:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19673 [014] 213577.939704:kvm_entry: vcpu 0
qemu-system-x86-19563 [010] 213577.939705:kvm_apic_accept_irq: apicid 0 vec 130 (LowPrio|level) qemu-system-x86-19332 [008] 213577.939706:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19563 [010] 213577.939707:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19332 [008] 213577.939713:kvm_mmio: mmio read len 4 gpa0xfed000f0val0x29a105deqemu-system-x86-19332 [008] 213577.939715:kvm_entry: vcpu 0
qemu-system-x86-19201 [011] 213577.939716:kvm_exit: reason exception rip0x1162412qemu-system-x86-19332 [008] 213577.939717:kvm_exit: reason halt rip0xfffffa6000fae7a1qemu-system-x86-19201 [011] 213577.939717:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939761:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939762:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939766:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939772:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f58ddqemu-system-x86-19673 [014] 213577.939774:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939776:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939776:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939779:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939782:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f5d09qemu-system-x86-19673 [014] 213577.939784:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939791:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939791:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939794:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939798:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f62fbqemu-system-x86-19673 [014] 213577.939799:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939802:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939802:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939805:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939808:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f66f4qemu-system-x86-19673 [014] 213577.939809:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939836:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939837:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939845:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19661 [014] 213577.939875:kvm_set_irq: gsi 11 level 1 source 0 qemu-system-x86-19661 [014] 213577.939876:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19661 [014] 213577.939876:kvm_apic_accept_irq: apicid 0 vec 130 (LowPrio|level) qemu-system-x86-19661 [014] 213577.939877:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19320 [008] 213577.939895:kvm_set_irq: gsi 11 level 1 source 0 qemu-system-x86-19320 [008] 213577.939896:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19320 [008] 213577.939897:kvm_apic_accept_irq: apicid 0 vec 130 (LowPrio|level) qemu-system-x86-19673 [014] 213577.939898:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f89f2qemu-system-x86-19320 [008] 213577.939899:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19673 [014] 213577.939900:kvm_inj_virq: irq 130
qemu-system-x86-19673 [014] 213577.939901:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939904:kvm_exit: reasonio_instructionrip0xfffffa6001acb435qemu-system-x86-19673 [014] 213577.939904:kvm_pio:pio_readat0xc033size 1 count 1 qemu-system-x86-19673 [014] 213577.939907:kvm_set_irq: gsi 11 level 0 source 0 qemu-system-x86-19673 [014] 213577.939907:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19673 [014] 213577.939908:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19673 [014] 213577.939910:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939912:kvm_exit: reasonapic_accessrip0xfffff800016a050cqemu-system-x86-19673 [014] 213577.939914:kvm_mmio: mmio write len 4 gpa0xfee000b0val0x0qemu-system-x86-19673 [014] 213577.939914:kvm_apic:apic_writeAPIC_EOI=0x0qemu-system-x86-19673 [014] 213577.939914:kvm_ack_irq: irqchip IOAPIC pin 11 qemu-system-x86-19673 [014] 213577.939915:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939918:kvm_exit: reasonext_irqrip0xfffff800016a12f0qemu-system-x86-19661 [014] 213577.939934:kvm_set_irq: gsi 11 level 1 source 0 qemu-system-x86-19661 [014] 213577.939935:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19661 [014] 213577.939936:kvm_apic_accept_irq: apicid 0 vec 130 (LowPrio|level) qemu-system-x86-19661 [014] 213577.939936:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19332 [008] 213577.939940:kvm_inj_virq: irq 130
qemu-system-x86-19332 [008] 213577.939941:kvm_entry: vcpu 0
qemu-system-x86-19332 [008] 213577.939944:kvm_exit: reasonio_instructionrip0xfffffa6000f9d435qemu-system-x86-19332 [008] 213577.939944:kvm_pio:pio_readat0xc033size 1 count 1 qemu-system-x86-19332 [008] 213577.939948:kvm_set_irq: gsi 11 level 0 source 0 qemu-system-x86-19332 [008] 213577.939949:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19332 [008] 213577.939949:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19673 [014] 213577.939950:kvm_inj_virq: irq 130
qemu-system-x86-19673 [014] 213577.939951:kvm_entry: vcpu 0
qemu-system-x86-19332 [008] 213577.939953:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939953:kvm_exit: reasonio_instructionrip0xfffffa6001acb435qemu-system-x86-19673 [014] 213577.939954:kvm_pio:pio_readat0xc033size 1 count 1 qemu-system-x86-19332 [008] 213577.939955:kvm_exit: reasonapic_accessrip0xfffff8000166e50cqemu-system-x86-19673 [014] 213577.939957:kvm_set_irq: gsi 11 level 0 source 0 qemu-system-x86-19332 [008] 213577.939958:kvm_mmio: mmio write len 4 gpa0xfee000b0val0x0qemu-system-x86-19673 [014] 213577.939958:kvm_pic_set_irq: chip 1 pin 3 (level|masked) qemu-system-x86-19332 [008] 213577.939958:kvm_apic:apic_writeAPIC_EOI=0x0qemu-system-x86-19673 [014] 213577.939958:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level) qemu-system-x86-19332 [008] 213577.939958:kvm_ack_irq: irqchip IOAPIC pin 11 qemu-system-x86-19332 [008] 213577.939959:kvm_entry: vcpu 0
qemu-system-x86-19332 [008] 213577.939961:kvm_exit: reasonept_violationrip0xfffff80001b3af8eqemu-system-x86-19332 [008] 213577.939961:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939962:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939964:kvm_exit: reasonapic_accessrip0xfffff800016a050cqemu-system-x86-19332 [008] 213577.939965:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939966:kvm_mmio: mmio write len 4 gpa0xfee000b0val0x0qemu-system-x86-19673 [014] 213577.939967:kvm_apic:apic_writeAPIC_EOI=0x0qemu-system-x86-19673 [014] 213577.939967:kvm_ack_irq: irqchip IOAPIC pin 11 qemu-system-x86-19673 [014] 213577.939968:kvm_entry: vcpu 0
qemu-system-x86-19332 [008] 213577.939969:kvm_mmio: mmio read len 4 gpa0xfed000f0val0x29a16a40qemu-system-x86-19332 [008] 213577.939971:kvm_entry: vcpu 0
qemu-system-x86-19332 [008] 213577.939981:kvm_exit: reasonept_violationrip0xfffff80001b3af8eqemu-system-x86-19332 [008] 213577.939982:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.939982:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939983:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19332 [008] 213577.939985:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939987:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19332 [008] 213577.939989:kvm_mmio: mmio read len 4 gpa0xfed000f0val0x29a1722fqemu-system-x86-19673 [014] 213577.939991:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8fae83qemu-system-x86-19332 [008] 213577.939991:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.939993:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.940010:kvm_exit: reasoncr_accessrip0xfffff800016ee2b2qemu-system-x86-19673 [014] 213577.940011:kvm_cr:cr_write4 =0x678qemu-system-x86-19673 [014] 213577.940017:kvm_entry: vcpu 0
Quoted 8 lines
qemu-system-x86-19673 [014] 213577.940019:
kvm_exit: reasoncr_accessrip0xfffff800016ee2b5qemu-system-x86-19673 [014] 213577.940019:kvm_cr:cr_write4 =0x6f8qemu-system-x86-19673 [014] 213577.940021:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.940048:kvm_exit: reason exception rip0xfffff800016a0620qemu-system-x86-19673 [014] 213577.940049:kvm_entry: vcpu 0
qemu-system-x86-19673 [014] 213577.940079:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.940080:kvm_page_fault: addressfed000f0error_code181 qemu-system-x86-19673 [014] 213577.940083:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0
Here is oprofile:
Quoted 27 lines
4117817 62.2029 kvm-intel.ko kvm-intel.ko vmx_vcpu_run 338198 5.1087 qemu-system-x86_64 qemu-system-x86_64 /usr/local/qemu/48bb360cc687b89b74dfb1cac0f6e8812b64841c/bin/qemu-system-x86_64 62449 0.9433 kvm.ko kvm.ko kvm_arch_vcpu_ioctl_run 56512 0.8537 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 copy_user_generic_string 52373 0.7911 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 native_write_msr_safe 34847 0.5264 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 schedule 34678 0.5238 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 fget_light 29894 0.4516 kvm.ko kvm.ko paging64_walk_addr 27778 0.4196 kvm.ko kvm.ko gfn_to_hva 24563 0.3710 kvm.ko kvm.ko x86_decode_insn 23900 0.3610 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 do_select 21123 0.3191 libc-2.10.90.so libc-2.10.90.so memcpy 20694 0.3126 kvm.ko kvm.ko x86_emulate_insn 19862 0.3000 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 kfree 19107 0.2886 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 __switch_to 18319 0.2767 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 update_curr 17981 0.2716 libc-2.10.90.so libc-2.10.90.so ioctl 17934 0.2709 librt-2.10.90.so librt-2.10.90.so clock_gettime 17874 0.2700 ioatdma.ko ioatdma.ko ioat2_issue_pending 17578 0.2655 libpthread-2.10.90.so libpthread-2.10.90.so pthread_mutex_lock 17041 0.2574 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 task_rq_lock 15806 0.2388 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 native_read_msr_safe 15292 0.2310 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 fput 14197 0.2145 libc-2.10.90.so libc-2.10.90.so memset 14167 0.2140 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 __up_read 13974 0.2111 kvm.ko kvm.ko kvm_arch_vcpu_put 13885 0.2097 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 select_task_rq_fair
Quoted 8 lines
13766 0.2079 bnx2.ko bnx2.ko bnx2_poll_work 13349 0.2016 kvm.ko kvm.ko find_highest_vector 13121 0.1982 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 __down_read 12518 0.1891 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 do_vfs_ioctl 12184 0.1840 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 try_to_wake_up 12095 0.1827 kvm.ko kvm.ko gfn_to_memslot 11870 0.1793 kvm.ko kvm.ko kvm_read_guest 11657 0.1761 tun.ko tun.ko tun_chr_aio_read
-Andrew
Quoted 18 lines · Andrew Theurer
I am running a windows workload which has 26 windows VMs running many instances of a J2EE workload. There are 13 pairs of an application server VM and database server VM. There seem to be quite a bit of
vm_exits, and it looks over a third of them aremmio_exit:Quoted 8 lines
efer_relo 0 exits 337139 fpu_reloa 247321 halt_exit 19092 halt_wake 18611 host_stat 247332 hypercall 0 insn_emul 184265Quoted 8 lines
insn_emul 184265 invlpg 0 io_exits 69184 irq_exits 52953 irq_injec 48115 irq_windo 2411 largepage 19 mmio_exit 123554
Quoted 1 line
I collected a kvmtrace, and below is a very small portion of that. Is there a way I can figure out what device the mmio's are for?
We want 'info physical_address_space' in the monitor.
Quoted 1 line
Also, is it normal to have lots of
ept_violations? This is a 2 socket Nehalem system with SMT on.
So long as pf_fixed is low, these are all mmio or apic accesses.
Quoted 7 lines
Quoted 7 lines
qemu-system-x86-19673 [014] 213577.939624:
kvm_page_fault: addressfed000f0error_code181
qemu-system-x86-19673 [014] 213577.939627:kvm_mmio: mmio
unsatisfied-read len 4 gpa0xfed000f0val0x0
qemu-system-x86-19673 [014] 213577.939629:kvm_mmio: mmio read len 4
gpa0xfed000f0val0xfb8f214d
hpet
Quoted 5 lines
Quoted 5 lines
qemu-system-x86-19673 [014] 213577.939631:
kvm_entry: vcpu 0 qemu-system-x86-19673 [014] 213577.939633:kvm_exit: reasonept_violationrip0xfffff8000160ef8e
qemu-system-x86-19673 [014] 213577.939634:kvm_page_fault: addressfed000f0error_code181
hpet - was this the same exit? we ought to skip over the emulated instruction.
Quoted 4 lines
Quoted 4 lines
qemu-system-x86-19673 [014] 213577.939693:
kvm_page_fault: addressfed000f0error_code181
qemu-system-x86-19673 [014] 213577.939696:kvm_mmio: mmio
unsatisfied-read len 4 gpa0xfed000f0val0x0
hpet
Quoted 6 lines
Quoted 6 lines
qemu-system-x86-19332 [008] 213577.939699:
kvm_exit: reasonept_violationrip0xfffff80001b3af8e
qemu-system-x86-19332 [008] 213577.939700:kvm_page_fault: addressfed000f0error_code181
qemu-system-x86-19673 [014] 213577.939702:kvm_mmio: mmio read len 4
gpa0xfed000f0val0xfb8f3da6
hpet
Quoted 6 lines
Quoted 6 lines
qemu-system-x86-19332 [008] 213577.939706:
kvm_mmio: mmio
unsatisfied-read len 4 gpa0xfed000f0val0x0
qemu-system-x86-19563 [010] 213577.939707:kvm_ioapic_set_irq: pin
11 dst 1 vec=130 (LowPrio|logical|level)
qemu-system-x86-19332 [008] 213577.939713:kvm_mmio: mmio read len 4
gpa0xfed000f0val0x29a105de
hpet ...
Quoted 7 lines
Quoted 7 lines
qemu-system-x86-19673 [014] 213577.939908:
kvm_ioapic_set_irq: pin
11 dst 1 vec=130 (LowPrio|logical|level)
qemu-system-x86-19673 [014] 213577.939910:kvm_entry: vcpu 0 qemu-system-x86-19673 [014] 213577.939912:kvm_exit: reasonapic_accessrip0xfffff800016a050c
qemu-system-x86-19673 [014] 213577.939914:kvm_mmio: mmio write len
4 gpa0xfee000b0val0x0
apic eoi
Quoted 6 lines
Quoted 6 lines
qemu-system-x86-19332 [008] 213577.939958:
kvm_mmio: mmio write len
4 gpa0xfee000b0val0x0
qemu-system-x86-19673 [014] 213577.939958:kvm_pic_set_irq: chip 1
pin 3 (level|masked)
qemu-system-x86-19332 [008] 213577.939958:kvm_apic:apic_writeAPIC_EOI=0x0
apic eoi
Quoted 7 lines
Quoted 7 lines
qemu-system-x86-19673 [014] 213577.940010:
kvm_exit: reasoncr_accessrip0xfffff800016ee2b2
qemu-system-x86-19673 [014] 213577.940011:kvm_cr:cr_write4 =0x678qemu-system-x86-19673 [014] 213577.940017:kvm_entry: vcpu 0 qemu-system-x86-19673 [014] 213577.940019:kvm_exit: reasoncr_accessrip0xfffff800016ee2b5
qemu-system-x86-19673 [014] 213577.940019:kvm_cr:cr_write4 =0x6f8
toggling global pages, we can avoid that with CR4_GUEST_HOST_MASK.
So, tons of hpet and eois. We can accelerate both by thing the hyper-V accelerations, we already have some (unmerged) code for eoi, so this should be improved soon.
Quoted 8 lines
Here is oprofile:
Quoted 5 lines
4117817 62.2029 kvm-intel.ko kvm-intel.ko vmx_vcpu_run 338198 5.1087 qemu-system-x86_64 qemu-system-x86_64 /usr/local/qemu/48bb360cc687b89b74dfb1cac0f6e8812b64841c/bin/qemu-system-x86_6462449 0.9433 kvm.ko kvm.ko kvm_arch_vcpu_ioctl_run 56512 0.8537 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 copy_user_generic_string
We ought to switch to put_user/get_user. rep movs has quite slow start-up.
Quoted 1 line
Quoted 1 line
52373 0.7911 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1
native_write_msr_safe
hpet in kernel or hyper-V timers will reduce this.
Quoted 2 lines
Quoted 2 lines
34847 0.5264 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 schedule 34678 0.5238 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 fget_light
and this.
Quoted 6 lines
Quoted 6 lines
29894 0.4516 kvm.ko kvm.ko paging64_walk_addr 27778 0.4196 kvm.ko kvm.ko gfn_to_hva 24563 0.3710 kvm.ko kvm.ko x86_decode_insn 23900 0.3610 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 do_select 21123 0.3191 libc-2.10.90.so libc-2.10.90.so memcpy 20694 0.3126 kvm.ko kvm.ko x86_emulate_insn
hyper-V APIC and timers will reduce all of the above (except memcpy).
Signature · 1 line
error compiling committee.c: too many arguments to function
hi,Avi:
I met the same problem, tons of hpet vm_exits(vector 209, fault
address is in the guest vm's hpet mmio range), even I disable hpet device in win7 guest vm, it still produce a larget amount of vm_exits when trace-cmd ; I add -no-hpet to start the vm, it still has HPET device inside VM.
Does that means the HPET device in VM does not depend on the
emulated hpet device in qemu-kvm? Is there any way to disable the VM HPET device to prevent so many vm_exits? Thansk.
Regards.
Suya.
2009/12/3 Avi Kivity <[email protected]>:
Quoted 171 lines · Andrew Theurer
Quoted 23 lines
I am running a windows workload which has 26 windows VMs running many instances of a J2EE workload. There are 13 pairs of an application server VM and database server VM. There seem to be quite a bit of
vm_exits, and it looks over a third of them aremmio_exit:Quoted 8 lines
efer_relo0
exits 337139fpu_reloa247321halt_exit19092halt_wake18611host_stat247332
hypercall 0insn_emul184265Quoted 8 lines
insn_emul184265
invlpg 0io_exits69184irq_exits52953irq_injec48115irq_windo2411
largepage 19mmio_exit123554Quoted 2 lines
I collected a kvmtrace, and below is a very small portion of that. Is there a way I can figure out what device the mmio's are for?
We want 'info
physical_address_space' in the monitor.Quoted 2 lines
Also, is it normal to have lots of
ept_violations? This is a 2 socket Nehalem system with SMT on.So long as
pf_fixedis low, these are all mmio or apic accesses.Quoted 8 lines
Quoted 8 lines
qemu-system-x86-19673 [014] 213577.939624:
kvm_page_fault: addressfed000f0error_code181
qemu-system-x86-19673 [014] 213577.939627:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19673 [014] 213577.939629:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f214dhpet
Quoted 5 lines
Quoted 5 lines
qemu-system-x86-19673 [014] 213577.939631:
kvm_entry: vcpu 0 qemu-system-x86-19673 [014] 213577.939633:kvm_exit: reasonept_violationrip0xfffff8000160ef8eqemu-system-x86-19673 [014] 213577.939634:kvm_page_fault: addressfed000f0error_code181hpet - was this the same exit? we ought to skip over the emulated instruction.
Quoted 4 lines
Quoted 4 lines
qemu-system-x86-19673 [014] 213577.939693:
kvm_page_fault: addressfed000f0error_code181
qemu-system-x86-19673 [014] 213577.939696:kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0hpet
Quoted 6 lines
Quoted 6 lines
qemu-system-x86-19332 [008] 213577.939699:
kvm_exit: reasonept_violationrip0xfffff80001b3af8e
qemu-system-x86-19332 [008] 213577.939700:kvm_page_fault: addressfed000f0error_code181
qemu-system-x86-19673 [014] 213577.939702:kvm_mmio: mmio read len 4 gpa0xfed000f0val0xfb8f3da6hpet
Quoted 6 lines
Quoted 6 lines
qemu-system-x86-19332 [008] 213577.939706:
kvm_mmio: mmio unsatisfied-read len 4 gpa0xfed000f0val0x0qemu-system-x86-19563 [010] 213577.939707:kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level)
qemu-system-x86-19332 [008] 213577.939713:kvm_mmio: mmio read len 4 gpa0xfed000f0val0x29a105dehpet ...
Quoted 7 lines
Quoted 7 lines
qemu-system-x86-19673 [014] 213577.939908:
kvm_ioapic_set_irq: pin 11 dst 1 vec=130 (LowPrio|logical|level)
qemu-system-x86-19673 [014] 213577.939910:kvm_entry: vcpu 0 qemu-system-x86-19673 [014] 213577.939912:kvm_exit: reasonapic_accessrip0xfffff800016a050c
qemu-system-x86-19673 [014] 213577.939914:kvm_mmio: mmio write len 4 gpa0xfee000b0val0x0apic eoi
Quoted 6 lines
Quoted 6 lines
qemu-system-x86-19332 [008] 213577.939958:
kvm_mmio: mmio write len 4 gpa0xfee000b0val0x0
qemu-system-x86-19673 [014] 213577.939958:kvm_pic_set_irq: chip 1 pin 3 (level|masked)
qemu-system-x86-19332 [008] 213577.939958:kvm_apic:apic_writeAPIC_EOI=0x0apic eoi
Quoted 7 lines
Quoted 7 lines
qemu-system-x86-19673 [014] 213577.940010:
kvm_exit: reasoncr_accessrip0xfffff800016ee2b2
qemu-system-x86-19673 [014] 213577.940011:kvm_cr:cr_write4 =0x678qemu-system-x86-19673 [014] 213577.940017:kvm_entry: vcpu 0 qemu-system-x86-19673 [014] 213577.940019:kvm_exit: reasoncr_accessrip0xfffff800016ee2b5
qemu-system-x86-19673 [014] 213577.940019:kvm_cr:cr_write4 =0x6f8toggling global pages, we can avoid that with
CR4_GUEST_HOST_MASK.So, tons of hpet and eois. We can accelerate both by thing the hyper-V accelerations, we already have some (unmerged) code for eoi, so this should be improved soon.
Quoted 13 lines
Here is oprofile:
Quoted 10 lines
4117817 62.2029 kvm-intel.ko kvm-intel.ko
vmx_vcpu_run
338198 5.1087 qemu-system-x86_64 qemu-system-x86_64/usr/local/qemu/48bb360cc687b89b74dfb1cac0f6e8812b64841c/bin/qemu-system-x86_6462449 0.9433 kvm.ko kvm.kokvm_arch_vcpu_ioctl_run
56512 0.8537 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1copy_user_generic_stringWe ought to switch to put_user/get_user. rep movs has quite slow start-up.
Quoted 4 lines
Quoted 4 lines
52373 0.7911 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1
native_write_msr_safehpet in kernel or hyper-V timers will reduce this.
Quoted 8 lines
Quoted 8 lines
34847 0.5264 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 schedule
34678 0.5238 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1fget_lightand this.
Quoted 6 lines
Quoted 6 lines
29894 0.4516 kvm.ko kvm.ko
paging64_walk_addr
27778 0.4196 kvm.ko kvm.kogfn_to_hva
24563 0.3710 kvm.ko kvm.kox86_decode_insnQuoted 8 lines
Quoted 8 lines
23900 0.3610 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1 vmlinux-2.6.32-rc7-5e8cb552cb8b48244b6d07bff984b3c4080d4bc9-autokern1
do_select
21123 0.3191 libc-2.10.90.so libc-2.10.90.so memcpy
20694 0.3126 kvm.ko kvm.kox86_emulate_insnhyper-V APIC and timers will reduce all of the above (except memcpy).
--
error compilingcommittee.c: too many arguments to function--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 8 lines · ya su
hi,Avi:
I met the same problem, tons of hpet vm_exits(vector 209, fault
address is in the guest vm's hpet mmio range), even I disable hpet device in win7 guest vm, it still produce a larget amount ofvm_exitswhen trace-cmd ; I add -no-hpet to start the vm, it still has HPET device inside VM.
Quoted 4 lines
Does that means the HPET device in VM does not depend on the
emulated hpet device in qemu-kvm? Is there any way to disable the VM HPET device to prevent so manyvm_exits? Thansk.
Looks like a bug to me.
Signature · 2 lines
I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.
Quoted 9 lines · Avi Kivity
Quoted 8 lines · ya su
hi,Avi:
I met the same problem, tons of hpet vm_exits(vector 209, fault
address is in the guest vm's hpet mmio range), even I disable hpet device in win7 guest vm, it still produce a larget amount ofvm_exitswhen trace-cmd ; I add -no-hpet to start the vm, it still has HPET device inside VM.
Quoted 6 lines
Quoted 4 lines
Does that means the HPET device in VM does not depend on the
emulated hpet device in qemu-kvm? Is there any way to disable the VM HPET device to prevent so manyvm_exits? Thansk.Looks like a bug to me.
IIRC disabling the HPET device doesn't remove the entry from the DSDT, no? So the guest OS might still think it's there while nothing responds (read returns -1).
Alex
HPET configuration in Seabios (was: Re: windows workload: many ept_violation and mmio exits)
Quoted 12 lines · Alexander Graf
Quoted 9 lines · Avi Kivity
Quoted 8 lines · ya su
hi,Avi:
I met the same problem, tons of hpet vm_exits(vector 209, fault
address is in the guest vm's hpet mmio range), even I disable hpet device in win7 guest vm, it still produce a larget amount ofvm_exitswhen trace-cmd ; I add -no-hpet to start the vm, it still has HPET device inside VM.
Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Does that means the HPET device in VM does not depend on the
emulated hpet device in qemu-kvm? Is there any way to disable the VM HPET device to prevent so manyvm_exits? Thansk.Looks like a bug to me.
IIRC disabling the HPET device doesn't remove the entry from the DSDT, no? So the guest OS might still think it's there while nothing responds (read returns -1).
Exactly. We have a fw_cfg interface in place for quite a while now (though I wonder how the firmware is supposed to tell -no-hpet apart from QEMU versions that don't provide this data - both return count = 255), but SeaBios still exposes one HPET block at a hard-coded address unconditionally.
There was quite some discussion about the corresponding Seabios patches back then but apparently no consensus was found. Re-reading it, I think Kevin asked for passing the necessary DSDT fragments from QEMU to the firmware instead of using a new, proprietary fw_cfg format. Is that still the key requirement for any patch finally fixing this bug?
Jan
Re: HPET configuration in Seabios (was: Re: windows workload: many ept_violation and mmio exits)
Quoted 28 lines · Jan Kiszka
Quoted 12 lines · Alexander Graf
Quoted 9 lines · Avi Kivity
Quoted 8 lines · ya su
hi,Avi:
I met the same problem, tons of hpet vm_exits(vector 209, fault
address is in the guest vm's hpet mmio range), even I disable hpet device in win7 guest vm, it still produce a larget amount ofvm_exitswhen trace-cmd ; I add -no-hpet to start the vm, it still has HPET device inside VM.Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Does that means the HPET device in VM does not depend on the
emulated hpet device in qemu-kvm? Is there any way to disable the VM HPET device to prevent so manyvm_exits? Thansk.Looks like a bug to me.
IIRC disabling the HPET device doesn't remove the entry from the DSDT, no? So the guest OS might still think it's there while nothing responds (read returns -1).
Exactly. We have a
fw_cfginterface in place for quite a while now (though I wonder how the firmware is supposed to tell -no-hpet apart from QEMU versions that don't provide this data - both return count = 255), but SeaBios still exposes one HPET block at a hard-coded address unconditionally.
Quoted 5 lines
There was quite some discussion about the corresponding Seabios patches back then but apparently no consensus was found. Re-reading it, I think Kevin asked for passing the necessary DSDT fragments from QEMU to the firmware instead of using a new, proprietary
fw_cfgformat. Is that still the key requirement for any patch finally fixing this bug?
My preference would be to use the existing ACPI table passing interface (fw_cfg slot 0x8000) to pass different ACPI tables to SeaBIOS.
SeaBIOS doesn't currently allow that interface to override tables SeaBIOS builds itself, but it's a simple change to rectify that.
When this was last proposed, it was raised that the header information in the ACPI table may then not match the tables that SeaBIOS builds. I think I proposed at that time that SeaBIOS could use the header of the first fw_cfg table (or some other fw_cfg interface) to populate the headers of its table headers. However, there was no consensus.
Note - the above is in regard to the HPET table. If the HPET entry in the DSDT needs to be removed then that's a bigger change.
-Kevin
Re: HPET configuration in Seabios
Quoted 42 lines · Kevin O'Connor
Quoted 28 lines · Jan Kiszka
Quoted 12 lines · Alexander Graf
Quoted 9 lines · Avi Kivity
Quoted 8 lines · ya su
hi,Avi:
I met the same problem, tons of hpet vm_exits(vector 209, fault
address is in the guest vm's hpet mmio range), even I disable hpet device in win7 guest vm, it still produce a larget amount ofvm_exitswhen trace-cmd ; I add -no-hpet to start the vm, it still has HPET device inside VM.Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Does that means the HPET device in VM does not depend on the
emulated hpet device in qemu-kvm? Is there any way to disable the VM HPET device to prevent so manyvm_exits? Thansk.Looks like a bug to me.
IIRC disabling the HPET device doesn't remove the entry from the DSDT, no? So the guest OS might still think it's there while nothing responds (read returns -1).
Exactly. We have a
fw_cfginterface in place for quite a while now (though I wonder how the firmware is supposed to tell -no-hpet apart from QEMU versions that don't provide this data - both return count = 255), but SeaBios still exposes one HPET block at a hard-coded address unconditionally.Quoted 5 lines
There was quite some discussion about the corresponding Seabios patches back then but apparently no consensus was found. Re-reading it, I think Kevin asked for passing the necessary DSDT fragments from QEMU to the firmware instead of using a new, proprietary
fw_cfgformat. Is that still the key requirement for any patch finally fixing this bug?My preference would be to use the existing ACPI table passing interface (
fw_cfgslot0x8000) to pass different ACPI tables to SeaBIOS.SeaBIOS doesn't currently allow that interface to override tables SeaBIOS builds itself, but it's a simple change to rectify that.
Quoted 9 lines
When this was last proposed, it was raised that the header information in the ACPI table may then not match the tables that SeaBIOS builds. I think I proposed at that time that SeaBIOS could use the header of the first
fw_cfgtable (or some otherfw_cfginterface) to populate the headers of its table headers. However, there was no consensus.Note - the above is in regard to the HPET table. If the HPET entry in the DSDT needs to be removed then that's a bigger change.
Can't seabios just poke at the hpet itself and see if it exists or not?
Signature · 2 lines
I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.
Re: HPET configuration in Seabios
Quoted 58 lines · Avi Kivity
Quoted 57 lines · Kevin O'Connor
Quoted 8 lines · Jan Kiszka
Quoted 7 lines · Alexander Graf
Quoted 4 lines · Avi Kivity
Quoted 3 lines · ya su
hi,Avi:
I met the same problem, tons of hpet vm_exits(vector 209,
fault
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
address is in the guest vm's hpet mmio range), even I disable
hpet
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
device in win7 guest vm, it still produce a larget amount of
vm_exitsQuoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
when trace-cmd ; I add -no-hpet to start the vm, it still has
HPET
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
device inside VM.
Does that means the HPET device in VM does not depend on the
emulated hpet device in qemu-kvm? Is there any way to disablethe VM
Quoted 6 lines
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines
HPET device to prevent so many
vm_exits? Thansk.Looks like a bug to me.
IIRC disabling the HPET device doesn't remove the entry from the
DSDT, no? So the guest OS might still think it's there while nothing responds (read returns -1).
Quoted 8 lines
Exactly. We have a
fw_cfginterface in place for quite a while now (though I wonder how the firmware is supposed to tell -no-hpet apart from QEMU versions that don't provide this data - both return count = 255), but SeaBios still exposes one HPET block at a hard-coded address unconditionally.There was quite some discussion about the corresponding Seabios
patches
Quoted 1 line
back then but apparently no consensus was found. Re-reading it, I
think
Quoted 3 lines
Kevin asked for passing the necessary DSDT fragments from QEMU to the firmware instead of using a new, proprietary
fw_cfgformat. Is that still the key requirement for any patch finally fixing this bug?My preference would be to use the existing ACPI table passing interface (
fw_cfgslot0x8000) to pass different ACPI tables to SeaBIOS.SeaBIOS doesn't currently allow that interface to override tables SeaBIOS builds itself, but it's a simple change to rectify that.
When this was last proposed, it was raised that the header information in the ACPI table may then not match the tables that SeaBIOS builds. I think I proposed at that time that SeaBIOS could use the header of the first
fw_cfgtable (or some otherfw_cfginterface) to populate the headers of its table headers. However, there was no consensus.
Quoted 6 lines
Quoted 3 lines
Note - the above is in regard to the HPET table. If the HPET entry in the DSDT needs to be removed then that's a bigger change.
Can't seabios just poke at the hpet itself and see if it exists or not?
Would be hard for the BIOS to guess the locations of the blocks unless we define the addresses used by QEMU as something like base + hpet_no * block_size in all cases.
Jan
Signature · 2 lines
Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux
Re: HPET configuration in Seabios
Quoted 8 lines · Jan Kiszka
Quoted 3 lines
Can't seabios just poke at the hpet itself and see if it exists or not?
Would be hard for the BIOS to guess the locations of the blocks unless we define the addresses used by QEMU as something like base +
hpet_no*block_sizein all cases.
Currently we have a fixed address. We could do:
if available in fw_cfg:
use that (may indicate no hpet)
elif fixed address works:
use that
else
no hpet
Signature · 1 line
error compiling committee.c: too many arguments to function
Re: HPET configuration in Seabios
Quoted 10 lines · Avi Kivity
Quoted 8 lines · Jan Kiszka
Quoted 3 lines
Can't seabios just poke at the hpet itself and see if it exists or not?
Would be hard for the BIOS to guess the locations of the blocks unless we define the addresses used by QEMU as something like base +
hpet_no*block_sizein all cases.
Quoted 8 lines
Currently we have a fixed address. We could do:
if available in
fw_cfg:
use that (may indicate no hpet)
elif fixed address works:
use that
else
no hpet
Currently, we also only have a single HPET block, but that's just because of some QEMU limitations that will vanish sooner or later. Then nothing will prevent multiple "-device hpet,base=XXX".
Jan
Signature · 2 lines
Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux
Re: HPET configuration in Seabios
Quoted 20 lines · Jan Kiszka
Quoted 10 lines · Avi Kivity
Quoted 8 lines · Jan Kiszka
Quoted 3 lines
Can't seabios just poke at the hpet itself and see if it exists or not?
Would be hard for the BIOS to guess the locations of the blocks unless we define the addresses used by QEMU as something like base +
hpet_no*block_sizein all cases.Quoted 8 lines
Currently we have a fixed address. We could do:
if available in
fw_cfg:
use that (may indicate no hpet)
elif fixed address works:
use that
else
no hpet
Quoted 4 lines
Currently, we also only have a single HPET block, but that's just because of some QEMU limitations that will vanish sooner or later. Then nothing will prevent multiple "-device hpet,base=XXX".
Yes, so we should enable the fw_cfg interface before that happens.
Signature · 1 line
error compiling committee.c: too many arguments to function
Re: HPET configuration in Seabios
In reply to Jan KiszkaQuoted 20 lines · Jan Kiszka
Quoted 10 lines · Avi Kivity
Quoted 8 lines · Jan Kiszka
Quoted 3 lines
Can't seabios just poke at the hpet itself and see if it exists or not?
Would be hard for the BIOS to guess the locations of the blocks unless we define the addresses used by QEMU as something like base +
hpet_no*block_sizein all cases.Quoted 8 lines
Currently we have a fixed address. We could do:
if available in
fw_cfg:
use that (may indicate no hpet)
elif fixed address works:
use that
else
no hpet
Quoted 3 lines
Currently, we also only have a single HPET block, but that's just because of some QEMU limitations that will vanish sooner or later. Then nothing will prevent multiple "-device hpet,base=XXX".
That said, some HPET probing (without any fw_cfg) may be a short-term workaround to fix Seabios until we defined The solution for communicating HPET block configurations.
Jan
Signature · 2 lines
Siemens AG, Corporate Technology, CT T DE IT 1 Corporate Competence Center Embedded Linux